shiveshnavin/pipelane-server
If you are the rightful owner of pipelane-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.
The Pipelane server is a platform that allows users to create custom pipelines for various automated tasks, integrating AI agents through a built-in Model Context Protocol (MCP) server.
Pipelane server
Users can create custom pipelines, such as running cron scheduled jobs or creating an Automated Image Post generation pipe. The MCP server is built into the platform to enable connection with AI agents. It provides a GraphQL API for managing pipelines and tasks, and serves the client-side application for a user interface to interact with the Pipelane Engine.
Features
- GraphQL API: Interact with Pipelanes, Pipetasks, and executions through a well-defined GraphQL schema.
- Pipeline Definition: Define pipelines with sequences of tasks and configure their execution flow.
- Task Types: Support for various built-in task types (e.g., API calls, delays, script evaluation, file operations).
- Scheduling: Schedule pipelines to run at specific times using cron expressions.
- Execution Management: Track pipeline and task executions, view logs, and monitor status.
- Client Application: Designed to work with a separate client application for a user interface.
- Express.js Integration: Easily integrate the Pipelane UI server into your existing Express.js applications.
Demo
Here's a custom pipelane built using the pipelane-server that executes the LinkedIn post generator using AI.
Here's a another custom pipelane for checking Stock market losses every 5 mins and notifying me over Call, WhatsApp, Telegram and FCM.
Main pipelane config Page.
Quick peek at the List Page.
Installation
To set up a express app and run the Pipelane server:
Clone the repository
cd pipelane-server
yarn install
cd client
npm run build
Run standalone
npm start
Or Host the app in your existing express js app
creatPipelaneServer(
variantConfig,
db
).then(pipelaneApp => {
app.use('/pipelane', pipelaneApp)
app.listen(...)
})
AI MCP Server
The pipelane tasks are also callable thorugh MCP.
Include below at the root (imp) of your experess app Install the below dependencies in your root project
npm i @modelcontextprotocol/sdk zod
import { createMcpServer } from "./server/mcp";
app.use(createMcpServer(VariantConfig, db))
SSL
Pass via env.
PIPELANE_HTTPS_PORT=8443
PIPELANE_HTTPS_KEY_PATH=/path/to/private.key.pem
PIPELANE_HTTPS_CERT_PATH=/path/to/certificate.pem
Notes
Firebase
Please create the indexes as well:
SQL
To support emojis in the output
ALTER TABLE ps_pipelane_task_exec
MODIFY output mediumtext
CHARACTER SET utf8mb4
COLLATE utf8mb4_unicode_ci;